Genome-wide identification of phytosulfokine (PSK) peptide family reveals TaPSK1 gene involved in grain development of wheat (Triticum aestivum L.)

Abstract

Background

Phytosulfokine (PSK) functions as a plant peptide growth factor that plays important and diverse roles in plant development and stress responses. Nevertheless, PSKs have not been systematically analyzed in wheat.

Results

A genome-wide comparative analysis of PSK genes in wheat was conducted and 15 TaPSKs were identified and divided into four subgroups in the wheat genome based on sequence similarity. The examination of motif compositions of TaPSK genes revealed the presence of the YIYTQ signature motif in the C-terminus of all TaPSK polypeptide precursors, with a highly conserved feature across different species. Exogenous application of TaPSK peptide promoted root growth in wheat. Quantitative RT-PCR analysis revealed that the TaPSKs exhibited preferential or tissue-specific expression patterns in wheat. In particular, three homologs of the TaPSK1 genes were specifically expressed in grains, with the strongest expression observed in the developing grains at 15 days after anthesis. Compared with wild type, transgenic rice lines overexpressing TaPSK1-A exhibited larger grain size and higher thousand-grain weight. The promoter region and genomic sequence of the wheat TaPSK1-A gene were cloned. Sequence polymorphism showed that five single-nucleotide polymorphisms (SNPs) were identified in the promoter region of TaPSK1-A. A Kompetitive Allele-Specific PCR (KASP) marker was developed for TaPSK1-A based on –806 bp SNP (C/T transition), and two haplotypes, TaPSK1-A-HapI and TaPSK1-A-HapII were detected in 260 wheat accessions collected from different regions. The expression of TaPSK1-A, promoter activity, and thousand-grain weight (TGW) in the TaPSK1-A-HapII haplotype were higher than those in the TaPSK1-A-HapI haplotype. Furthermore, yeast one-hybrid assays revealed the binding of TaNF-YB1 and TaERF39 to the promoter regions of the TaPSK1-A gene, and TaMADS29 could bind to the promoters of TaPSK1-B and TaPSK1-D genes.

Conclusions

Comparative genome-wide analysis of TaPSK peptide family revealed that the TaPSK1 gene is involved in wheat grain development, and the developed TaPSK1-A-KASP marker could be utilized for marker-assisted selection breeding of wheat.

Graphical Abstract

Introduction

Signaling peptides play crucial regulatory roles in plant development and symbiotic interactions, and are involved in defense responses. Over the past decades, many studies have revealed that secreted peptides act as short- and long-distance signaling molecules in cell-to-cell communication and exert their functions by interacting with specific receptors [1, 2]. In plants, 43 types of peptides have been identified, and the total number of small peptides is increasing [3]. Peptides are generally defined as proteins consisting of 2 to 100 amino acid residues and are encoded by small open reading frames that contribute to plant growth and development [4].

Phytosulfokine (PSK) is a peptide growth factor widely distributed in higher plants. PSK was originally described to act in the proliferation of low-density cell suspension [5]. Subsequent studies demonstrated that the addition of chemically synthesized PSK peptide can stimulate the proliferation of asparagus single mesophyll cells and delay senescence [6]. Previous studies have shown that the low-affinity PSK can bind to its membrane-localized receptor PSKR1 to activate the downstream signaling pathway, which is involved in many biological processes in plants (Stuhrwohldt et al. 2011).

PSK is identified as a five-amino-acid peptide (Tyr(SO3H)-Ile-Tyr(SO3H)-Thr-Gln) belonging to the sulfated plant peptide group [7]. It is derived from 80–120 amino acid prepropeptides encoded by a small gene family [8]. The PSK homologs share a highly conserved pentapeptide sequence, YIYTQ, located near the C-terminus of the precursor peptides [3]. The tyrosine sulfation of the PSK precursor is catalyzed by the tyrosylprotein sulfotransferase (TPST) localized in the cis-Golgi, a unique gene in Arabidopsis [9], and then processed by processing enzymes in the apoplast [10].

In recent years, the PSK peptide has been identified in various plant species, such as Oryza sativa [11], Arabidopsis thaliana [12], Zea mays [13], and Pyrus bretschneideri [14]. Early studies have demonstrated that PSK-α promotes cell expansion and elongation in plants. Many studies have shown that PSK peptide participates in a variety of biological processes, including organ development [15], pollen tube elongation [14], somatic embryogenesis [16, 17], and adventitious root formation, as well as having important functions in plant immunity [18]. For example, in Arabidopsis, overexpression of the PSK-α precursor gene significantly increases the growth rate of leaves, roots, and hypocotyls. AtPSK1 and AtPSK3 are highly expressed in root tip cells [19]. AtPSK1 affects root cell elongation by regulating the size of mature cells. AtPSK2 is involved in the regulation of pollen tube growth and guidance, indicating that members of the PSK family also play a crucial role in reproductive development [20]. AtPSK4 is involved in the regulation of root and leaf development. Overexpression of Cunninghamia lanceolata ClPSK1 or ClPSK2 promotes root growth and adventitious root formation in Arabidopsis thaliana [21]. GhPSK promotes cotton fiber cell elongation by regulating K⁺ efflux, resulting in longer and finer fibers [22]. In maize (Zea mays L.), ZmPSK1 and ZmPSK3 show high expression in cells of female and male gametophytes [13]. Overexpression of TaPSK5 significantly enhances primary root growth, leading to an increase in the grain yield of rice. PbrPSK2 has been reported to regulate pear pollen tube elongation [14]. Apart from its role in regulating plant growth and development, PSK is also reported to be involved in regulating plant immune response in Arabidopsis and tomato (Solanum lycopersicum). In tomato, the PSK peptide increases cytosolic Ca²⁺ levels and triggers immune responses in plants against Botrytis cinerea by activating an auxin-dependent immune pathway [23].

At present, the biological and physiological functions of PSK precursor genes mainly focused on Arabidopsis thaliana. However, the systematic studies on the PSK peptide family in wheat are still very limited. Given the critical roles of PSK peptide in plant growth, development, and immune responses, in this study, we identified 15 TaPSK genes in wheat at the genome-wide level and investigated their evolutionary relationships, gene structures, conservative motifs, cis-acting regulatory elements, duplication events as well as the tissue expression patterns of TaPSKs. Notably, the TaPSK1-A gene was specifically expressed in grains at 15 DAA. We explored the function of the TaPSK1-A gene in rice and found that the transgenic lines displayed an increase in grain size and thousand-grain weight. We isolated TaPSK1-A and discovered its polymorphism sites for developing molecular marker. Furthermore, association analysis was performed to study favorable allelic variation of TaPSK1-A with grain-related traits among a natural population of common wheat accessions. The study also aimed to identify the transcription factors binding to the promoter regions of TaPSK1-A/B/D genes. Our results hold significant importance for further studies of TaPSK genes and offer valuable genetic resources for wheat breeding.

Materials and methodology

Identification of the TaPSK gene family

To identify the putative TaPSK genes, the full-length sequences of PSK proteins from Arabidopsis and rice were used to perform a BLASTP search against the WheatOmics 1.0 database (http://202.194.139.32/). The identified potential hits meet the following criteria: expected E-values of < 20, the propeptide sequences ranging from 60 and 200 amino acids with a signature motif in the C-terminal half. Wheat PSK protein sequences were downloaded from the Ensembl Plants database (http://plants.ensembl.org/index.html). All putative TaPSK proteins were aligned with the Arabidopsis and rice PSK proteins, and those containing the YIYTQ domain were defined as TaPSK proteins, respectively.

TaPSK protein features analysis

ExPASy (https://web.expasy.org/compute_pi/) was utilized to analyze the molecular weight (kDa) and the theoretical isoelectric point (pI) of each TaPSK protein. TargetP 2.0 (http://www.cbs.dtu.dk/services/SignalP/) was employed to predict the N-terminal signal peptide of the PSK proteins in wheat.

Gene structure and conservative motifs of TaPSK gene family

The predicted coding sequences (CDS) and genomic sequences of wheat PSK family members were downloaded from the Ensembl Plants database (http://plants.ensembl.org/index.html). Then the intron/exon distributions of the TaPSK genes were performed using the Gene Structure Display Server 2.0 based on annotation information (GSDS 2.0, https://gsds.cbi.pku.edu.cn/index.php). The PSK motifs were performed using the Multiple Expectation Maximization for Motif Elicitation (MEME) suite with default settings (http://meme-suite.org/tools/meme).

Analysis of cis-acting elements in TaPSK genes

The wheatOmics 1.0 database (http://202.194.139.32/) was utilized to retrieve the upstream sequence (2000 bp) of the initiation codon (ATG) of each TaPSK gene. Subsequently, these sequences were analyzed for cis-regulatory elements using the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). The promoter cis-elements were visualized via the GSDS 2.0 (http://gsds.cbi.pku.edu.cn/).

Chromosomal locations, sequence alignment, and phylogenetic analysis

The wheat genome annotation file was obtained from the Ensembl Plants database (http://plants.ensembl.org/info/data/ftp/index.html). The chromosomal locations of TaPSK genes were then mapped on chromosomes using TBtools software [24]. Multiple sequence alignments of PSK protein sequences from wheat, rice, and Arabidopsis were conducted using Clustal function in MEGA 7.0. The phylogenetic tree was performed using the neighbor-joining method with 1,000 bootstrap replicates. The EvolView website (https://evolgenius.info//evolview-v2/#login) was used to optimize the display of the phylogenetic tree.

Gene duplication

MCScanX [25] was used to identify duplicated gene pairs of wheat TaPSK genes. The gene duplication events of TaPSK genes meet the following two criteria: (1) the similarity of two alignment sequences had an identity > 80%, and (2) the alignment length covered over 80% of the longer aligned sequence [26]. The duplicated regions between paralogous gene pairs were illustrated using the Circos function in TBtools software [27]. KaKs_Calculator 2.0 software [28] was employed to analyze the non-synonymous substitution (Ka) and synonymous (Ks) values of the duplicated gene pairs.

Expression patterns analysis of TaPSKs by RNA-seq data

To further analyze the expression pattern of TaPSK genes, RNA-Seq data from five tissues (root, stem, leaf, spike, and grain) were obtained from the Wheat Expression Browser (http://202.194.139.32/expression/wheat.html) and used to investigate the expression patterns of TaPSK genes in wheat. The TPM (transcripts per million mapped reads) values were calculated for each TaPSK gene and a heat map was generated using TBtools software [24].

Plant material and growth conditions

For analyses of tissue-specific expression patterns, samples from roots, stems, and leaves at the three-leaf stage, as well as spikes at the flowering stage and grains at 5, 10, 15, 20, and 25 days after anthesis (DAA) were collected from Chinese Spring wheat plants. All samples were immediately frozen in liquid nitrogen and stored at − 80 °C for subsequent RNA extraction.

Expression analysis of TaPSK genes

Total RNA was extracted using the RNAprep pure plant total RNA extraction kit (TIANGEN, Beijing, China). The first-strand cDNA was synthesized using the ReverTra Ace qPCR RT Master Mix with gDNA Remover kit (TOYOBO, Japan). Quantitative real-time PCR was performed using the KOD SYBR qPCR Mix (TOYOBO, Japan) with QuanStudio™ 5 Real-Time PCR software (Foster City, CA, USA). Each 20 µL reaction mixture contained 2 × KOD SYBR qPCR Mix 10 μL, cDNA 1 μL, forward and reverse primers (10 μmol/L) 0.4 μL for each, and ddH₂O 8.2 μL. The qPCR reaction procedure was carried out as follows: 95 ℃ for 2 min, followed by 40 cycles of 98 ℃ for 10 s, 60 ℃ for 15 s and 68 ℃ for 20 s. All qPCR analyses were repeated three times in this study. TaActin was used as the internal reference gene and the relative gene expression was calculated with the 2^−ΔΔCt method [29].

Synthesis of the TaPSK peptide and application

The five amino acids of the TaPSK (Tyr(SO3H)-Ile-Tyr(SO3H)-Thr-Gln) peptide were synthesized by the DGpeptide company (http://www.dgpeptides.com/) with a purity exceeding 90%. The wheat seeds were sterilized in 70% ethanol for 30 s and then washed with distilled water for five times. Afterward, the seeds were sterilized with 30% bleach for 10 min, followed by four to five times rinses with distilled water. Sterilized seeds were sown on 0.5 $\times$ Murashige and Skoog (MS), 0.5% agar, and 1% sucrose for germination. When the seeding roots reached a length of about 0.5 cm, the seeds were removed to avoid nutrient interference. Twenty seedlings were then transferred to new plates supplied with 500 nM TaPSK peptide, 0.5 $\times$ MS, 0.5% agar, and varying sucrose concentrations of 0%, 0.3%, and 0.6%. The seedling roots were measured after 6 days of incubation in the growth chamber with a 14 h light/10 h dark cycle, 25/20 °C temperature, and 60% relative humidity.

Transgenic vector construction, plant transformation, and microscopy analysis

The full-length cDNA of TaPSK1-A was amplified from the grains of wheat variety Fielder and then inserted into the KpnI and BamHI sites of the overexpression vector UbiBGV005-CCDB, driven by the ubiquitin promoter. The TaPSK1-A overexpression vector was transformed into the japonica rice variety zhonghua11 using Agrobacterium tumefaciens-mediated transformation. For scanning electron microscopy, the outer surfaces of mature grains of wild type and transgenic rice plants were photographed using a scanning electron microscope (S-4800, Hitachi).

Haplotype-based association analysis and evolution analysis

The 1.7-kb promoter sequence (upstream of ATG) of TaPSK1-A was extracted from the WheatOmics 1.0 database [30]. Twenty diverse wheat cultivars with significantly different grain weights were used to detect the SNPs of the TaPSK1-A gene. A pair of primers was used to amplify the promoter region of TaPSK1-A. The nucleotide polymorphisms for TaPSK1-A in wheat varieties were identified using the SeqMan package of Lasergene 7.1 software (DNASTAR, USA). After sequencing the PCR products, five SNPs were found among these genotypes. A KASP marker was developed to discriminate the two haplotypes at TaPSK1-A. The KASP primers were developed based on the SNP (C/T transition) at base pair position –806 by following the standard procedure. Two allele-specific forward primers were designed with the fluorophore FAM and HEX tails attached to the SNP at the 3′ end. Each forward primer and one common reverse primer were designed. The KASP assays were carried out according to the KASPline protocol (https://www.lgcgroup.com) and analyzed using KLUSTERCALLER genotyping software (LGC Biosearch Technologies, Hoddesdon, Herts, UK). The developed marker TaPSK1-A-KASP was then applied across a natural population of 260 winter wheat accessions. The frequency and geographic distribution of two TaPSK1-A haplotypes in China were determined using 374 varieties, including 151 varieties from 260 varieties used for association analysis and 223 varieties accessed from WheatUnion. The effects of haplotype on the yield-related traits were analyzed by SPSS 19.0 software (SPSS Corp, Chicago, IL, USA).

Dual-luciferase assay

The 1700-bp promoter regions of haplotypes TaPSK1-A-HapI and TaPSK1-A-HapII of TaPSK1-A were amplified from the Chinese wheat cultivars Yumai2 and Jimai44, respectively, and then fused in the pGreen II 0800 vector controlled by the CaMV35 S promoter. The resulting recombinant plasmids, TaPSK1-A-HapI-LUC and TaPSK1-A-HapII-LUC, were transformed into the Agrobacterium tumefaciens strain EHA105. The promoter activity was determined using a Dual-Luciferase Reporter Assay System (Vazyme, China) according to the manufacturer's manual. The Firefly luciferase (LUC)/ Renilla luciferase (REN) ratio was used for determining promoter activity.

Yeast one-hybrid assays

The full-length CDS sequences of TaNF-YB1 (TraesCS6B02G316700), TaERF39 (TraesCS1B02G233300), and TaMADS29 (TraesCS6D02G147400) were amplified and separately inserted into the pGADT7 vector. Fragments of the TaPSK1-A promoter (base pairs –1129 to –980 and –1271 to –1173), the TaPSK1-B promoter (base pairs –411 to –214), and the TaPSK1-D promoter (base pairs –371 to –183) were amplified and cloned into the pHIS2 vector. The primers used for amplifying these transcription factors and promoter fragments are listed in Additional file 1: Table S1. The pairs of combined pGADT7-TaNF-YB1 and pHIS2-TaPSK1-A-pro1, pGADT7-TaERF39 and pHIS2-TaPSK1-A-pro2, pGADT7-TaMADS29 and pHIS2-TaPSK1-B-pro3, and pGADT7-TaMADS29 and pHIS2-TaPSK1-D-pro4 plasmids were co-transformed into yeast strain Y187 according to the manufacturer's instructions. The vector pGADT7 was co-transformed with each reporter vector as negative controls, while the pGAD53 and pHIS2-p53 vectors were used as positive control. Yeast colonies were grown on SD/-Leu/-Trp medium at 30 ℃ for 3 days and then spotted onto SD/-Leu/-Trp/-His medium containing 25 mM 3-AT to assess interactions.

Results

Genome-wide identification of TaPSK gene family

To identify TaPSK genes in wheat, the full-length PSK protein sequences of Arabidopsis and rice, along with the HMM PSK domain profile PF02536, were used to search the PSK homologs in the wheat genomic database. Subsequently, the candidate protein sequences were confirmed through multiple sequence alignment. As a result, 15 putative TaPSK genes were identified in the wheat genome and designated as TaPSK1-A to TaPSK5-D based on their genomic homology. The chromosome location of the wheat TaPSK gene family revealed that 15 TaPSKs were unevenly distributed on 12 wheat chromosomes, which were assigned to 4 homologous gene groups. Among them, six TaPSK genes are found in homoeologous group 4 chromosomes, while homoeologous group 2, 5, and 7 chromosomes each contained only one TaPSK gene. There are 5 homeo-paralogs in each of the A, B, and D sub-genomes (Fig. 1A).

Fig. 1

Chromosome localization, phylogenetic relationship, and genomic organization of TaPSK genes. A The chromosome location of TaPSK genes in wheat. B Phylogenetic relationships of 29 PSK the precursor peptides from O. sativa (7), A. thaliana (7) and T. aestivum (15). The different color branches represent different groups of the PSK family. C Gene structure in the TaPSK family members. The blue boxes, black lines, and yellow boxes indicate the untranslated (UTRs), introns, and exons, respectively

The detailed characteristics of the identified TaPSK proteins were summarized, including the number of amino acids, molecular weight, isoelectric points, and the presence of signal peptides (Additional file 1: Table S2). The TaPSK proteins ranged from 86 (TaPSK1-A/B) to 128 (TaPSK4-D) amino acid residues in length, with predicted molecular weights ranging from 9,055.28 Da (TaPSK1-A) to 13,352.06 Da (TaPSK4-D). The isoelectric points varied from 4.5 (TaPSK3-D) and 9.1 (TaPSK5-D), with 86.7% (13/15) of the TaPSK proteins having an isoelectric point less than 7, indicating that these TaPSK proteins are rich in acidic amino acids. Furthermore, the signal peptides of all 15 TaPSK family members were predicted using TargetP 2.0 and iPSORT, revealing signal peptide sequence lengths ranging from 25 to 28 amino acids (Additional file 1: Table S2).

Multiple alignment and phylogenetic analysis of TaPSK proteins

All full-length sequences of 15 TaPSK proteins were aligned using Clustal W. As shown in Additional file 1: Figure S1, all proteins contained the conserved sequence of the mature peptide (DYIYTQ). TaPSK precursor proteins within homologous gene groups exhibited higher similarity, while those in non-homologous gene groups showed much lower homologous similarity. To gain further insights into the evolutionary relationships of TaPSK proteins, we selected 29 PSK protein sequences from three species (7 members from A. thaliana, 7 members from rice, and 15 members from wheat), and constructed a phylogenetic tree based on the alignment of the full-length protein sequences via the neighbor-joining method by using MEGA7.0. The result showed that PSK gene members were categorized into four groups (Group I, II, III, and IV), containing 10, 9, 5, and 5 PSK genes, respectively (Fig. 1B). Group I comprised six AtPSKs, one OsPSK and three TaPSKs. Notably, 85.7% (6/7) of the AtPSK genes were clustered in group I, indicating higher sequence similarity among these paralogs compared to ortholog proteins from wheat and rice. Group III contained two OsPSKs and three TaPSKs, while genes in groups II and III were not present in A. thaliana. Group IV was composed of one AtPSK, one OsPSK, and three TaPSKs.

Gene structure and motif analysis of TaPSKs

The wheat TaPSK genes clustered into the same group exhibited similar exon–intron structures, conserved motifs, and domain structures. Structural analysis revealed that five TaPSK genes (TaPSK2-B/D and 3-A/B/D) possessed one exon, while the remaining TaPSK genes had two exons and one intron (Fig. 1C). Additionally, we analyzed the conserved motifs in TaPSK proteins using the MEME online server. The results showed that three homologous copies within the same clade had similar conserved motifs, distinct from those found in separate subgroups. The conserved motifs 1, 2, and 3 were identified in all TaPSK proteins. The DYIYTQ domain was represented by the conserved motif 3, while motif 1 corresponded to the signal peptide. The three homologous copies of TaPSK2 and TaPSK3 contained six motifs, which was more than the number of motifs found in proteins of other subfamilies (Additional file 1: Figure S2).

Duplication patterns of TaPSK genes

To investigate the duplication events of TaPSK genes, MCScanX was used to identify tandem and segmental duplication events, and their syntenic relationships were visualized using Tbtools software. In terms of gene duplication, all wheat PSK members were found to be involved in segmental duplication, which contained 15 segmental duplication gene pairs in the TaPSK gene family. Furthermore, five homologous gene groups with a copy on each of the A, B, and D sub-genomes were identified as segmental duplication genes (Additional file 1: Figure S3). These results indicated that segmental duplication events were responsible for the expansion of the TaPSK gene family. To better understand these duplicated gene pairs, we calculated the ratio of Ka (Nonsynonymous) and Ks (Synonymous). The Ka/Ks ratios of the 15 pairs of homoeologous genes ranged from 0.066 to 1.70, with an average ratio of 0.551 (Additional file 1: Table S3). A pair of duplicated genes, TaPSK3-A and TaPSK3-D, had a ratio greater than 1, indicating that they are under positive selection, while the remaining duplicated gene pairs had ratios less than 1, suggesting that these segmentally duplicated TaPSKs may have undergone purifying selection.

Cis-acting elements in the promoters of TaPSK genes

The cis-acting elements in the 2-kb upstream genomic sequence of all TaPSK genes were identified using the PlantCARE database. This study evaluated hormone-responsive elements, stress-responsive elements, and growth and development-related elements, as shown in Additional file 2: Table S1. ABA-responsive elements such as ABRE, and the methyl jasmonate (MeJA)-responsive elements including CGTCA-motif and TGACG-motif were found in most TaPSK genes. All three TaPSK2 homoeologous genes contained ABRE3a and ABRE4. In addition, IAA response elements including AuxRR-core and TGA-element, salicylic acid elements such as TCA-element, and gibberellin (GA) elements like P-box were distributed in the promoter regions of certain wheat TaPSK genes. For stress-responsive elements, MYC and MYC were present in the promoter regions of all TaPSKs. STRE (stress-responsive element) was found in 93% (15) of TaPSK genes. Moreover, multiple abiotic stress response elements were widely distributed among different TaPSK genes, such as drought response element MBS (MYB binding site), dehydration-responsive element (DRE), and WRKY binding site (W-box). The low temperature-responsive element (LTR) was observed in the promoters of five TaPSKs (TaPSK4-A/B/D, 1-B, and 1-D) (Additional file 2: Table S1).

Expression profiles of TaPSKs in tissues

To reveal the potential roles of TaPSK genes in growth and development, we analyzed their expression profiles in 5 tissues at different growth stages using the available RNA-seq data from the Wheat Expression Browser database. The results showed that 15 wheat TaPSK genes exhibited tissue-specific or predominant expression patterns. Notably, TaPSK1-A, 1-B, and 1-D showed the highest expression levels in grain_Z71, while TaPSK2-A, 2-B, and 2-D were predominantly expressed in leaf_Z23 and leaf_Z71. TaPSK3-B and 3-D displayed high expression levels during panicle and grain development. TaPSK3-A, TaPSK5-A, and TaPSK5-D were mainly expressed in grain_Z85. TaPSK4-A, 4-B, and 4-D exhibited high expression levels in stem_Z32 and spikes at three developmental stages. TaPSK5-B had dominant transcript levels in root_Z13 and root_Z39 (Fig. 2).

Fig. 2

Heatmap of wheat TaPSK gene expressions in 5 different tissues. The RNA-sequence data were obtained from root, stem, leaf, spike, and grain. The data used in the figure were obtained from expVIP. Z10: one-leaf stage; Z13: three-leaf stage; Z23: early tiller stage; Z30: booting stage; Z32: early jointing stage; Z39: Late jointing stage; Z65: mid-flowering stage; Z71: 2 d after flowering; Z75: 10 d after flowering; Z85: 30 d after flowering. Heat map was generated using log2-transformed TPM expression values. Color green or red indicates low to high expression abundance, respectively

To validate the transcriptome data, seven selected genes as representatives were detected by qRT-PCR. The results showed that TaPSK1-A and 1-B were strongly expressed in grains at different growth stages, with the highest expression levels at 15 DAA. However, their expression in other tissues was almost undetectable. The expression levels of TaPSK2-A, 2-B, and 2-D were significantly higher in leaves compared to other tissues. Meanwhile, the TaPSK5-B gene transcript was predominantly expressed at 25 DAA in grains (Fig. 3).

Fig. 3

qRT-qPCR assessment of the expression levels of selected wheat TaPSK genes in different tissues, including roots, stems, leaves, spikes at flowering, and grains at 5, 10, 15, 20, and 25 days after anthesis (DAA). Data are the mean ± standard errors of three independent replicates

TaPSK peptide promotes wheat root growth

We synthesized the TaPSK peptide to evaluate its physiological effect on wheat root growth. After the seeds germinated, the seeds were removed and wheat seedlings with similar primary root lengths were transferred to new agar plates supplied with 500 nM TaPSK peptide and varying concentrations of sucrose. After another 6 days in the plant growth chamber, the length of the primary roots was measured. The results indicated that exogenously applied TaPSK peptide significantly promoted wheat root growth (Additional file 1: Figure S4). When exposed to different sucrose concentrations, both root fresh weight and root length were significantly greater in the TaPSK peptide treatment than those of wild-type controls (Additional file 1: Figure S4).

Fig. 4

TaPSK1-overexpressing rice lines exhibit larger seeds. A Phenotypic comparison of grains at the adult stage between wild-type (WT) and TaPSK1 overexpression line (OE). Bar = 1 cm. Comparisons of thousand-grain weight (B) and grain length (C) of TaPSK1 transgenic rice and the wild-type Zhonghua11 (ZH11) at the mature stage. D Scanning electron microscopy analysis of epidermal cells in lower parts of WT and TaPSK1-OE3 grains. Bar, 100 μm. The cell length E and cell width F of the mature grain. Values are the mean ± SD. *p < 0.05, **p < 0.01

TaPSK1-A overexpression increased grain size in Oryza sativa

Since TaPSK1-A was specifically expressed in grain tissue, we selected TaPSK1-A for further investigation of its function. To test the effect of TaPSK1-A on grain development, the full-length TaPSK1-A coding sequence was cloned and inserted into the UbiBGV005-CCDB overexpression vector. Subsequently, three transgenic lines with higher overexpression levels were selected for further study. The transgenic plants with overexpressed of TaPSK1-A displayed increased thousand-grain weight, grain length, and larger grain size compared to the wild type (Fig. 4A–C). Electron microscopy of the outer epidermal cells of mature grains revealed that the overexpressed lines had larger outer epidermal cells than those of the wild type (Fig. 4D), suggesting that TaPSK1-A plays a crucial role in regulating grain size development. In addition, the length of epidermal cells in the overexpression line was significantly greater than those in the wild type, while there was no significant difference in cell width between overexpressed lines and wild type plants (Fig. 4E, F). The findings demonstrated that TaPSK1-A promotes cell elongation in the seed coat, leading to increased the length and grain weight of mature seeds.

Association and evolutionary analysis of TaPSK1-A in cultivated wheat

To identify the sequence polymorphisms of TaPSK1-A, -B, and -D genes, SNPs in the promoters and coding regions were analyzed using genomic sequence data from 627 accessions obtained from the WheatUnion database (http://wheat.cau.edu.cn/WheatUnion/). The results showed that no SNP differences were identified in the promoters and coding regions of TaPSK1-B and TaPSK1-D. However, five single-nucleotide polymorphisms were detected in the promotor region of TaPSK1-A by sequencing 20 cultivated wheat accessions, with no SNPs identified in its coding region (Fig. 5A, B). Two haplotypes were identified for TaPSK1-A based on sequence variants in natural populations. To distinguish these haplotypes, a KASP marker was developed based on the SNP at –806 bp upstream of the TaPSK-1A coding region (Fig. 5B, C).

Fig. 5

Nucleotide polymorphism of development of a molecular marker of TaPSK1-A haplotype analysis. A Five single-nucleotide polymorphisms (SNPs) were identified in the promoter region of TaPSK1-A. B A KASP marker was developed based on the -806 bp (C/T). C KASP assay for SNP-806 nt allele “T” in FAM and allele in “C” HEX cluster. Varieties exhibiting blue dots have the FAM-type allele, and red-colored dots have the HEX-type allele. TaPSK1-A haplotypes associated with D thousand-grain weight, and E grain length across three environments (E1, E2, E3). E1 to E3 indicate the environments of Tongwei (2019–2020), Tongwei (2020–2021), and Zhuanglang (2021–2022), respectively. *P<0.05, **P<0.01, respectively. Error bars indicate the standard error 

To detect the effect of haplotypes and agronomic traits, a population consisting of 260 wheat accessions was used for the association analysis. Among wheat populations screened by the KASP genotyping platform, two haplotypes, TaPSK1-A-HapI and TaPSK1-A-HapII, were observed in 73 and 187 cultivars, respectively (Additional file 1: Table S4). The association analysis revealed that the cultivars with TaPSK1-A-HapII exhibited significantly higher thousand-grain weight (TGW) and longer grain length compared to those with TaPSK1-A-HapI across three environments (E1, E2, and E3) (Fig. 5D, E). Therefore, the TaPSK1-A-HapII (T) was considered as a superior allele for TGW and grain length.

To analyze the effect of favorable allelic variation on grain weight, 12 wheat accessions possessing TaPSK1-A-HapI and TaPSK1-A-HapII alleles were selected for examining the expression levels of TaPSK1-A (Fig. 6A). The results showed from qRT-PCR indicated that the genotypes with TaPSK1-A-HapII had higher relative expression levels than those with TaPSK1-A-HapI in grains during the middle stage of grain development. This implies that the TaPSK1-A gene may be positively correlated with higher thousand-grain weight in wheat.

Fig. 6

Expression pattern and promoter activity analysis of TaPSK1-A haplotype. A Relative expression levels of the wheat TaPSK1-A haplotypes TaPSK1-A-HapI and TaPSK1-A-HapII in grains at 15 days after anthesis. The first six genotypes, Yumai 2, Taishan 1, Longmai 847, Changzhi 516, Zhongmai 533, and Yunhan 2129 possess the allelic variation TaPSK1-A-HapI. The second six genotypes, Jimai 22, Jimai 30, Zhongyou 9507, Jimai 44, Zhoumai 18, and Chang 4738 possess the allelic variation TaPSK1-A-HapII. Error bars denote ± SE. B The activity of TaPSK1-A promoters with haplotype TaPSK1-A-HapI and TaPSK1-A-HapII. The two haplotype promoters were fused with the reporter vector pGreenII 0800-LUC. The LUC/REN ratio represents the relative activity of the TaPSK1-A-HapI and TaPSK1-A-HapII promoters

Given the presence of SNPs in the promoter region of TaPSK1-A haplotypes, it was hypothesized that genetic variability may alter its promoter activity. Therefore, we evaluated the effects of two TaPSK1-A promoter haplotypes on promoter activity. The results revealed that the promoter activity from haplotype TaPSK1-A-HapII was higher than that of haplotype TaPSK1-A-HapI (Fig. 6B). This suggested that the increased promoter activity of TaPSK1-A may be involved in maintaining higher expression levels, and ultimately contributing to the variation in grain weight.

Geographic distribution of TaPSK1-A haplotypes in the Chinese wheat production zones

The geographic distribution of two haplotypes was evaluated among the 374 modern cultivars from 11 provinces in China (Additional file 1: Table S5). As shown in Fig. 7, the TaPSK1-A-HapI haplotype was predominant in Sichuan (68%), Xinjiang (75%), Tibet (76%), and Yunnan (87%). However, in the main wheat growing regions of Shandong, Henan, Hebei, Shanxi, and Jiangsu, the frequencies of TaPSK1-A-HapII were higher than that of TaPSK1-A-HapI. Thus, favorable allelic variation of TaPSK1-A-HapII was selected in the major wheat production areas in the process of wheat breeding in China (Fig. 7).

Fig. 7

Geographical distribution of the two TaPSK1-A haplotypes in China. Geographic distribution of TaPSK1-A haplotypes in 374 modern cultivars. The map was downloaded in the Standard Map Service System (http://bzdt.ch.mnr.gov.cn/)

Yeast one-hybrid screening verification of three potential TFs binding to the promoters of TaPSK1-A/B/D

To further analyze the transcription factors binding to the promoters of TaPSK1-A/B/D genes, we initially utilized the wGRN database (http://wheat.cau.edu.cn/wGRN/) for predicting transcription factor binding sites. The database predicted that three transcription factors, including the ethylene-responsive transcription factor ERF39, nuclear transcription factor Y subunit B-1-like, and MADS-box transcription factor 29-like isoform X2, wound bind to the promoter regions of TaPSK1-A/B/D. To validate this prediction, we performed a yeast one-hybrid assay in which the full-length CDS sequences of TaNF-YB1, TaERF39, and TaMADS29 were cloned into the pGADT7 vector. In addition, short fragments of the TaPSK1-A/B/D promoters, named TaPSK1-A-pro1, TaPSK1-A-pro2, TaPSK1-B-pro3, and TaPSK1-D-pro4, were fused to the pHIS2 reporter gene. These combined plasmids pGADT7-transcription factors + pHIS2-promoters, pGADT7 + pHIS2-promoters, and pGAD53 + pHIS2-p53 were then co-transformed into yeast Y187 competent cells and cultured on SD/–Trp/–Leu/–His/3-AT yeast medium. The results of the yeast one-hybrid assays confirmed that TaNF-YB1 and TaERF39 can bind to the TaPSK1-A promoter region, while TaMADS29 bound to the promoters of TaPSK1-B and TaPSK1-D. This suggests that these transcription factors can bind to their specific binding sites and regulate the expression of TaPSK1-A, TaPSK1-B, and TaPSK1-D (Fig. 8).

Fig. 8

Yeast one-hybrid assay to test the interaction between transcription regulators and the promoter regions of TaPSK1-A/B/D. Yeast one-hybrid results for three transcription factors binding to the fragments of TaPSK1-A/B/D promoters in wheat. TaNF-YB1, TaERF39, and TaMADS29 were fused with the activation domain (AD). TaPSK1-A-pro1, TaPSK1-A-pro2, TaPSK1-B-pro3, and TaPSK1-D-pro4 represent the promoter regions of TaPSK1-A/B/D fused to the pHIS2 reporter gene. pGAD53m and pHIS-p53 were used as the positive control. The concentrations of 3-Amino-1,2,4-Triazole (3-AT) are shown at the top. The transformants were grown on SD/-Leu-Trp, SD/-Leu-Trp-Leu and SD/-Leu-Trp-Leu + 3-AT, respectively. SD/-Leu-Trp is the synthetic dropout (SD) medium supplemented without Leu and Trp, and SD/-Leu-Trp-Leu is the SD medium supplemented without Leu, Trp, and His. AD: pGADT7. BD: pGBKT7

Discussion

The PSK peptide plays key roles in the growth and development of plants. The PSK gene family has been identified and characterized in several species so far [15, 21, 31]. However, this gene family has not been systematically investigated in wheat. In this study, we identified 15 TaPSK genes in wheat, a number larger than that in other previously estimated organisms, such as Arabidopsis (Arabidopsis thaliana, seven members) [3], maize (Zea mays, four members) [13], rice (Oryza sativa, seven members) [32], and tomato (Solanum lycopersicum, eight members) [31]. Compared to other large gene families, the members of this peptide family in different plant species were encoded by a relatively small gene family.

Phylogenetic analysis revealed that the 15 wheat PSK genes were divided into four subfamilies, with group II further divided into two subfamilies. A comparison between wheat and rice PSK peptide families showed an even distribution across different subgroups, indicating that a higher homology between them, and the genetic relationship of wheat and rice PSK genes are more closely related than those of Arabidopsis thaliana. In contrast, most of the PSK genes in Arabidopsis clustered together on a single branch, highlighting dissimilarities between wheat and Arabidopsis PSKs. Based on gene structures and motif analysis, wheat PSKs in the same clade exhibited similar motif compositions and gene structures, implying functional similarities among proteins in the same evolutionary branch. Multiple sequences alignment showed that the PSK prepropeptide sequences of wheat, rice, and Arabidopsis differ greatly at the N-terminal, except for AtPSK6 which contained the pentapeptide YIYTH, however, they all shared the conserved sequence YIYIQ at the C-terminus of the polypeptide. The expression level of AtPSK6 was mostly undetectable, indicating that AtPSK6 may be a pseudogene [33].

Previous studies have demonstrated the crucial role of sulfation on two tyrosine residues (YIYTQ) for PSK activity and function. Some studies have indicated that sulfonylation of these tyrosine residues can increase PSK bioactivity by approximately 1000-fold and enhance its binding affinity with PSKR1 receptor [3]. The TPST enzyme has been identified as being responsible for the sulfation of certain peptides, including root meristem growth factor 1 (RGF1), PSK, CLE-Like (CLEL) families, and plant peptides containing sulfated tyrosine (PSY) [34, 35]. The PSK precursor polypeptide containing the tyrosine site (-EDYGD-) was catalyzed by the Golgi-localized protein AtTPST [36]. In addition, 15 TaPSK precursor proteins share an Asp residue at the –1 position of the pentapeptide, which is identical to all PSK peptide homologs in tomato and Arabidopsis [31, 37]. However, the C-terminal amino acid residues downstream of the scissile bond (in position P1′) were not completely conserved (Additional file 1: Figure S1). Previous studies have shown that exogenously applied 1 µm PSK peptide resulted in approximately a 20% longer primary root compared to untreated plants [19]. Our study found that treating wheat seedlings with 500 nmol PSK peptide increased both the primary root length and root fresh weight, especially in the presence of a low concentration of sucrose. These results indicate that PSK-α promotes root growth in wheat, with a similar effect observed in Arabidopsis roots.

The functions of PSK genes in Arabidopsis thaliana have been elucidated in recent years. PSK peptide is involved in the regulation of cell proliferation [2]. Although all PSK genes in different species encode mature peptides containing YIYTQ or YIYTH, their functions vary in plant development. To dissect the expression patterns of TaPSK genes in wheat, qRT-PCR was conducted to detect their expression patterns in different tissues. The results showed that TaPSK2-A/B/D exhibited the highest expression levels in leaves, with no expression detected in seeds, implying a potential role in wheat leaf growth. On the other hand, TaPSK1-A/B/D genes were specifically expressed in developing seeds, indicating that TaPSK1 may play a crucial role in seed development.

Previous reports have revealed that the TaPSK5 gene, referred to as TaPSK4 in this study, plays a crucial role in regulating root growth and grain yield in rice [38]. In this study, the function of the TaPSK1 was chosen for detailed investigation due to its high expression and specificity in developing seeds. The polymorphisms of TaPSK1 gene sequences were extracted and analyzed from the WheaUnion database (http://wheat.cau.edu.cn/WheatUnion/b_4/) [39]. Subsequently, SNPs in the 1.7-kb promoter region of the TaPSK1-A gene were identified in 20 wheat accessions through direct sequencing. The functional marker was developed based on variations in the TaPSK1-A promoter region to distinguish the two haplotypes in the germplasm population. Further analysis of haplotype and grain weight revealed that accessions possessing TaPSK1-A-HapII had higher TGW, and the relative expression levels of TaPSK1 were higher in wheat genotypes with TaPSK1-1A (T) than those with (C). This finding was consistent with the result that overexpression of TaPSK1-A can increase grain size and weight in rice. Furthermore, there were differences in promoter activities between the two haplotypes, suggesting that SNP variation in the promoter region may be accountable for the observed phenotypic traits and gene expression. Previous studies have shown that polymorphisms in gene promoter regions can lead to an increase in thousand kernel weight or grain size [40]. For example, allelic variation in the promoter region of TaCYP78A5 leads to changes in promoter activity and expression levels, thereby affecting thousand-grain weight between the two haplotypes. Overall, these results suggest that the TaPSK1-A gene is involved in the regulation of grain size and development, and the utilization of SNPs in the superior haplotype of TaPSK1-A could be a promising strategy for marker-assisted selection breeding in wheat.

Nuclear Factor Y (NF-Y) transcription factor NF-YB1 plays important roles in regulating plant growth and development [41, 42]. The NF-YB1 orthologs in rice, known as OsNF-YB1, are endosperm-specific genes and are involved in grain filling and influence endosperm cell proliferation [41,42,43]. Knock-down of NF-YB1 leads to a significantly delay in grain filling and results in smaller grain size. Another study revealed that OsNF-YB1 binds to the OsYUC11 promoter, and activates the expression of auxin biosynthesis genes to boost rice grain size [44]. Previous studies in other crop species have shown that the MADS29 transcription factor regulates grain development [45]. OsMADS29 has been found to regulate seed development by affecting the programmed cell death (PCD) of maternal tissues [46], as well as embryo and endosperm development through influencing cytokinin biosynthesis [45]. Recently, TaMADS29 has been observed to interact with TaNF-YB1 to regulate grain development, and overexpression of TaMADS29 increased grain width and grain weight [47]. In rice, overexpression of OsERF115 enhances grain-filling activity, affecting rice grain weight and grain size [48]. OsNF-YB1 has been identified to interact with OsERF115 to regulate endosperm development. In this current study, TaNF-YB1 (the ortholog of OsNF-YB1), TaMADS29 (the ortholog of OsMADS29), and TaERF39 were selected to investigate their binding as transcription factors. Yeast one-hybrid assay results showed that TaNF-YB1 and TaERF39 were found to bind to the promoter regions of TaPSK1-A, while TaMADS29 interacted with the promoters of TaPSK1-B and TaPSK1-D. It is proposed that these transcription factors regulate the transcription of TaPSK1-A/B/D to promote endosperm cell proliferation, ultimately affecting wheat grain filling and thousand-grain weight. The impact of these transcription factors on TaPSK1 expression requires further verification by other approaches such as electrophoretic mobility shift assay (EMSA). The results of this study provide a potential mechanism for understanding three TaPSK1 homoeologous genes involving in grain development.

Conclusions

In this study, 15 TaPSK genes were identified in T. aestivum at the genome-wide level. All TaPSK precursor peptides contain the highly conserved YIYTQ signature motif in the C-terminus. Exogenous application of TaPSK peptide promoted root growth in wheat. qRT-PCR analysis revealed that TaPSK1-A gene was specifically expressed in developing seeds. Five SNPs were identified in the promoter regions of TaPSK1-A. A molecular marker TaPSK1-A-KASP was developed based on the SNP at -806 bp (T-C). Association analysis revealed that the favorable allelic variation TaPSK1-A (T) was related to the traits of TGW and grain length. Overexpression of the TaPSK1-A gene in rice increased grain size and thousand-grain weight. Additionally, yeast one-hybrid assays demonstrated that three potential transcription factors, TaNF-YB1, TaERF39, and TaMADS29, could bind to the promoter regions of TaPSK1-A/B/D. Taken together, these findings offer valuable insights into the molecular regulation of TaPSK1 involved in seed development in wheat, and also provide the newly developed molecular marker, TaPSK1-A-KASP, which holds potential for marker-assisted selection in breeding programs aimed at improving the TGW of wheat.